AN INNOVATIVE NUMERICAL APPROACH FOR TRAIN PASS-BY NOISE FORECASTING

This paper deals with an engineering method for the prediction of vehicle pass-by noise based on a FEM/ BEM exterior acoustic calculation in the frequency domain. The researchers simulate, in a virtual environment, the experimental outdoor pass-by noise measurement. The simulated pass-by noise campaign is synthesized from multiple acoustic transfer functions between a line of virtual microphones located 7.5m on the side of the vehicle and each noise source. A numerical FEM/BEM train bogie acoustic model has been created within the MSC ACTRAN commercial softwares. Wheel-rail rolling noise, engine and powertrain noise acoustic source have been implemented and posi-tioned inside the FEM and BEM model to demonstrate the validity of the proposed methods. The contribu-tion from noise sources, expressed both in terms of sound pressure level and overall value, to the pass-by noise were evaluated up to 5 kHz. The virtual pass-by-noise assessment has been then validated by experi-mental measurement of the complete four coach’s train with respect to different speed regimes.